• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文学位 >Adaptive software rejuvenation.
【24h】

Adaptive software rejuvenation.

机译：自适应软件复兴。

获取原文
获取原文并翻译 | 示例
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

In this dissertation, we present a framework for adaptive software rejuvenation (ASR). Software rejuvenation is a proactive fault management technique. Its objective is to counter the phenomenon of software aging, the gradual accumulation of potential fault conditions and their impacts during system operation. Software rejuvenation has gained quick recognition since it emerged in 1995, and has broader applications as software systems have become ubiquitous in today's economic, social, medical, and other infrastructures. This dissertation addresses itself to the new, pressing, and challenging issue of adapting software rejuvenation to the dynamic changes of a system. A typical software system undergoes many changes over its lifetime, most notable in system components, resources, services, and client requirements, among others. System changes are the rule instead of the exception. The purpose of adaptive software rejuvenation (ASR) is to maintain system availability, reliability or other measures while a system undergoes the aging process and dynamic changes.; Adaptive software rejuvenation poses some challenges, particularly with respect to theoretical understanding and practical implementation. The ASR framework presents a basic, integrated solution. It includes system modeling, online inspection, adaptive degradation estimation, and adaptive rejuvenation scheduling. We introduce new concepts and methodologies pertaining to all these aspects of the framework. In modeling, we have unified the modeling techniques to enable the traditional analysis of systems in ideal conditions, the incorporation of online measured data that pertains to the aging process, and the adaptive analysis of systems in degradation. In the degradation estimation, we provide a methodology to isolate the effects of the aging process, particularly resource leaks, from the other failure factors. This is essential to avoid adverse software rejuvenation. We provide a methodology to incorporate measured data with a degradation model to provide an adaptive degradation estimate. In rejuvenation scheduling, we have built a degradation-and-rejuvenation hierarchy with a rejuvenation model on top of a degradation model. We have introduced a method for providing adaptive software strategies to maintain a system under expected conditions or reach its highest potential in certain aspect. The ASR methodologies designed to realize the ASR objective are not limited to those introduced and illustrated in this dissertation. They may be more application-specific in practice or more advanced in general.
机译：本文提出了一种适应性软件复兴的框架。软件复兴是一种主动的故障管理技术。其目的是解决软件老化现象，潜在故障条件的逐步积累及其在系统运行期间的影响。自从1995年出现以来，软件复兴已经得到了迅速的认可，并且随着软件系统在当今的经济，社会，医疗和其他基础架构中变得无处不在，它具有更广泛的应用。本文致力于解决使软件回春适应系统动态变化的新的，紧迫而富挑战性的问题。典型的软件系统在其生命周期中会发生许多变化，其中最值得注意的是系统组件，资源，服务和客户端需求等。系统更改是规则，而不是例外。自适应软件复兴（ASR）的目的是在系统经历老化过程和动态变化时维持系统可用性，可靠性或其他措施。自适应软件的复兴带来了一些挑战，特别是在理论理解和实际实施方面。 ASR框架提供了一个基本的集成解决方案。它包括系统建模，在线检查，自适应降级估计和自适应复兴计划。我们介绍与框架所有这些方面有关的新概念和方法。在建模中，我们统一了建模技术，以使传统系统能够在理想条件下进行传统分析，并纳入与老化过程有关的在线测量数据，并对退化中的系统进行自适应分析。在降级估算中，我们提供了一种方法来将老化过程的影响（尤其是资源泄漏）与其他故障因素隔离开。这对于避免不利的软件复兴至关重要。我们提供了一种将测量数据与退化模型合并在一起的方法，以提供自适应的退化估计。在复兴计划中，我们在退化模型之上建立了一个具有复兴模型的退化和复兴层次结构。我们已经介绍了一种提供自适应软件策略的方法，以在预期条件下维护系统或在某些方面发挥最大潜力。为实现ASR目标而设计的ASR方法不仅限于本文介绍和说明的方法。它们在实践中可能更特定于应用程序，或者总体上更高级。

著录项

  • 作者

    Bao, Yujuan.;

  • 作者单位

    Duke University.;

  • 授予单位 Duke University.;
  • 学科 Computer Science.
  • 学位 Ph.D.
  • 年度 2004
  • 页码 235 p.
  • 总页数 235
  • 原文格式 PDF
  • 正文语种 eng
  • 中图分类 自动化技术、计算机技术;
  • 关键词

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号